Cork Constraint Computation Centre (Profile)

Difficult problems can offer too many choices, many of which are incompatible, few of which are optimal. The Cork Constraint Computation Centre (4C) develops the basic science that will make it easier for computers to help us make these choices. Some examples of constraints: • The meeting must start at 6:30. • The separation between the soldermasks and nets should be at least 0.15mm. • This model only comes in blue and green. • This cable will not handle that much traffic. • These sequences should align optimally. • John prefers not to work on weekends. • The demand will probably be for more than 5 thousand units in August. Some examples of constraint satisfaction or optimization problems: • Schedule these employees to cover all the shifts. • Optimize the productivity of this manufacturing process. • Configure this product to meet my needs. • Find any violations of these design criteria. • Optimize the use of this satellite camera. • Align these amino acid sequences. Constraints arise in design and configuration, planning and scheduling, diagnosis and testing, and in many other contexts. Constraint programming can solve problems in telecommunications, internet commerce, electronics, bioinformatics, transportation, network management, supply chain management, and many other fields. Constraint computation has seen fundamental scientific advances, e.g. in understanding the relationship between problem structure and problem complexity. Constraint technology has demonstrated its commercial value. Research Agenda We apply advances in artificial intelligence and other disciplines to make constraint programming more powerful, more practical and easier to use. The work is centred in the field of artificial intelligence, but embedded in the broader constraint programming community. The centre conducts basic research in areas vital to the next generation of constraint technology. The process of modeling domain knowledge, tailoring heuristics, and exploring alternatives must become more automated. Specific topics include acquisition, validation, optimisation, learning, and explanation. Progress can be made here by abstracting our experience with specific applications. Applications: Applications will motivate and validate advances in automation. Application domains may include bioinformatics, configuration, computer and telecommunications networks, design, electronic commerce, planning and scheduling. Adaptability: In the real world we are confronted with uncertainty and change, with probabilities and preferences, with failures and tradeoffs, with collaborators and competitors. New applications and new contexts, e.g. interactive internet applications, present new challenges. Constraint technology must be further enriched to cope better with these challenges.